11. Question: At what clock speeds does HyperTransport(TM) technology operate?
Answer: HyperTransport(TM) technology devices are designed to operate at multiple clock speeds from 200MHz up to 800MHz, and utilizes double data rate technology transferring two bits of data per clock cycle, for an effective transfer rate of up to 1,600Mb/sec in each direction. Since transfers can occur in both directions simultaneously, an aggregate transfer rate of 6.4 Gigabytes per second in a 16 bit HyperTransport(TM) I/O Link and an aggregate transfer rate of 12.8 Gigabytes per second in a 32-bit HyperTransport(TM) I/O Link can be achieved. To allow for system design optimization, the clocks of the receive and transmit links may be set at different rates.
----
For the pentium4:
133MHz Quad Pumped (533MHz effective) allowing access to up to 4.2GB Bandwidth
But I guess that most of the trafic is mem-hd or hd-mem and thus does not nead to go trough the cpu, I think the latest alphadesign was to have 8 rambus chanels giving plenty of bandwith
If its not optical! Then every fiber transmits data in parallel! But its no use at the moment cause the electric2optical circuts is adding a lot of latency so for short distance it is no use, at least for now.
DRAM modules (or the mem-controller) does not lie on the pci-buss. It lies on the northbridge or on the cpu itself (like hammer). Nothing is forcing you to place the SCSI controller on the pci-buss.
I think they have a 64-bit version of UT, Read it on slashdot maby (a do not know), and that it gave some 10% speed increase from the 32-bit version (both running on an hammer). I do not think it gained any increase in speed with 64-bit types, the increase came from the that more registers are avaible in 64-bit mode.
How many of of the computers in www.top500.org is macs? *zero* They are IBM:s with Power3/4 with huge caches, often dual cores. They do *not* exist in Macs and are so fast mainly because they exist in fast interconnect networks, i.e. supercomputers. Macs powerPC:s are made by motorola and are *slow*.
Allmost... Bananas do have flowers, but they are defect and hence can not reproduce sexualy.
Bannanas are reproduced by taking the small plants comming up from the main plant i.e. they are cloned. The only changes that occure is therefore by mutations.
They can be used for processes as well, but the real advantage is when used by two threads belonging to the same process, then it will not trash the caches.
Slashdot Mirror
There is a diffrence between documentation and software, a good documentation license need not
be good for software, and vice versa.
So I think the GNU documentation license is ok for
documentation, and should be allowed for
documentation, but NOT for software.
from:
http://www.hypertransport.org/
11.
Question:
At what clock speeds does HyperTransport(TM) technology operate?
Answer:
HyperTransport(TM) technology devices are designed to operate at multiple clock speeds from 200MHz up to 800MHz, and utilizes double data rate technology transferring two bits of data per clock cycle, for an effective transfer rate of up to 1,600Mb/sec in each direction. Since transfers can occur in both directions simultaneously, an aggregate transfer rate of 6.4 Gigabytes per second in a 16 bit HyperTransport(TM) I/O Link and an aggregate transfer rate of 12.8 Gigabytes per second in a 32-bit HyperTransport(TM) I/O Link can be achieved. To allow for system design optimization, the clocks of the receive and transmit links may be set at different rates.
----
For the pentium4:
133MHz Quad Pumped (533MHz effective) allowing access to up to 4.2GB Bandwidth
But I guess that most of the trafic is mem-hd
or hd-mem and thus does not nead to go trough
the cpu, I think the latest alphadesign was
to have 8 rambus chanels giving plenty of
bandwith
If its not optical! Then every fiber transmits
data in parallel! But its no use at the moment
cause the electric2optical circuts is adding a
lot of latency so for short distance it is no
use, at least for now.
On a hyper-transport link or on the northbridge
like the agp buss.
DRAM modules (or the mem-controller) does not
lie on the pci-buss. It lies on the northbridge
or on the cpu itself (like hammer). Nothing is
forcing you to place the SCSI controller on the
pci-buss.
No, but the great boost is comming from the
increase in register count. FP arithmetic
is allready greater then 64-bit.
I think they have a 64-bit version of UT,
Read it on slashdot maby (a do not know), and that
it gave some 10% speed increase from the 32-bit
version (both running on an hammer). I do not
think it gained any increase in speed with 64-bit
types, the increase came from the that more
registers are avaible in 64-bit mode.
>Ok, then back to the JAR example: the program
>doesn't require the GPLed JAR, it just requires
>any JAR that happens to implement that interface.
Right.
>Based upon this, wouldn't any software linking
>against libraries (assuming they GPL-ed) on a >system be required to be GPL-ed?
Yes.
>Seems like it would be difficult to write any >software that didn't need to be GPL-ed if you >were doing it on a Linux system.
Not realy. Glibc is GPL with an exception that permits dynamic linking for any program.
Gtk is under LGPL, X is under a BSD-like license
and so on...
Just don't try to link readline!
>... forming a new combined work ... must be
>distributed under the terms of the GPL and only
>the GPL.
I.e. your code can be distributed under BSD,
but the combined work (BSD + GPL) must be under
the GPL.
Othervise you could unGPL code by adding a bit
of code under the BSD licence.
>But can the "trusted computing" scenarion beused
>to protect GNU works?
No.
>Can we make sure that others are not infringing
>on the GPL using this or am I grasping at straws
>here?
No, you are grasping at straws here.
>Is there anything in this initiative that could
>be used to promote our freedoms?
No.
1) recording nuclear breakdown
2) record the key on a cdrom and deliver
it by hand.
well x86 is fastest.
But why did the use Xeons????
They will not preform better, they will only
be more costly. Make it 2048 with normal p4.
I would think "repeat until people start to
beieve it" works for Apple too.
When will the Apple mouses have more than
one button? Do not answer that you can plug
any usb mouse.
It lets you---and enforces you to---write
invalid c++ code! g++ with --ansi stays
(pretty well) to the standard.
How many of of the computers in www.top500.org
is macs? *zero* They are IBM:s with Power3/4
with huge caches, often dual cores. They do
*not* exist in Macs and are so fast mainly because
they exist in fast interconnect networks, i.e.
supercomputers. Macs powerPC:s are made by motorola and are *slow*.
Allmost... Bananas do have flowers, but they
are defect and hence can not reproduce sexualy.
Bannanas are reproduced by taking the small
plants comming up from the main plant i.e.
they are cloned. The only changes that occure
is therefore by mutations.
They can be used for processes as well, but
the real advantage is when used by two threads
belonging to the same process, then it will not
trash the caches.
Yes, as said in the other posts,
BUT, you want to schedule the
same process on the same CPU in
order to not trash the cache.
I.e. you can make a huge inprovement
by make the scheduler aware of
processors *and* logical processors.
It DOES exist in windows
But I still can not get a grip of the mouse!!!
(or keyboard)
No, it is not fair.
Sell the printer for what it is worth.
Do you want to buy a car and then not be able to
buy parts from other companies???
The device driver is a part of the OS and not the hardware...
n (13056 * lines in loop)
The US should be ashamed that it can not produce
students of good quality, NOTHING else.